giraudier, s.; Baranda, A. F.; Bansaye, V.; Lauret, E.; Mounier, M.; Ugo, V.; Meleard, S.; Giraudier, S.

Mathematical modelling allows us to better understand the emergence and evolution of myeloproliferative neoplasms. We tested different mathematical models on a first cohort (patients) (Cote d\'Or Registry) to determine the onset and evolution times before JAK2V617F classical myeloproliferative disorders (polycythemia vera and essential thrombocythemia) are diagnosed. We considered the time to diagnosis as the sum of two periods: the time (from embryonic development) for the JAK2V617F mutation to appear, not disappear and enter proliferation, and a second period corresponding to the expansion of the clonal population until diagnosis. Using increasingly complex models, we show that the rate of active mutation cannot be constant, but rather increases exponentially with age, following the well-known Gompertz model. We found that it takes an average of 63.1 +/- 13 years for the first tumor cell to appear and start proliferating. On the other hand, the expansion time is constant: 8.8 years once the mutation has occurred. These results were validated in an external cohort (national FIMBANK cohort). Using this model, we analyzed JAK2V167F Essential Thrombocythemia versus Polycythemia Vera and found that the time to active mutation in PV is about 1.5 years longer than in ET, while the expansion time is similar. In conclusion, our multi-step approach and the final age-dependent model for the onset and development of MPN shows that the onset of a JAKV617F mutation should be linked to an ageing mechanism and indicates a period of 8-9 years for the development of a full MPN.